The search for potential genetic and environmental causes of schizophrenia began not long after Eugen Bleuler coined the name of the disease in 1908. In the first decades of the 20th century, twin and adoption studies began to consistently point to an inherited component in the susceptibility to the disease. In the decades before the molecular era, the twin data implied that it was a multigenic disease, with multiple genes of modest or small effect.

That general sense of a multifactorial etiology became apparent at a time during which evidence for an environmental component came with surprising difficulty. That evidence emerged slowly from careful epidemiologic studies of second-trimester influenza infections, and from the consequences of deliberate starvation of the Dutch population by the Nazis during the winter of 1944. After both events, women who endured these conditions during their second trimesters had more babies who grew up to have schizophrenia. Researchers have also observed that obstetric and perinatal problems are associated with the disease, and they now generally accept that physical problems during pregnancy can increase the likelihood of schizophrenia.

As the molecular era began in the middle of the 20th century, researchers confirmed the likely multigenic nature of schizophrenia's etiology. Reseachers began applying the first human molecular genetic maps to schizophrenia soon after such maps became available in the 1980s. They have since linked certain chromosomal regions more or less consistently to schizophrenia in different studies, and some genes in those regions appear to be associated with susceptibility.

Norton, Williams, and Owen, in a 2006 review, concluded that "the strongest evidence for putative schizophrenia susceptibility loci relates to the genes encoding dysbindin (DTNBP1) and neuregulin (NRG1). For other genes, disrupted in schizophrenia (DISC1), D-amino acid oxidase activator (DAOA), regulator of G-protein signaling 4 (RGS4) and V-AKT murine thymoma viral oncogene homolog 1 (AKT1), the data are promising but not yet compelling. In the most convincing cases, the risk haplotypes appear to be associated with small effect sizes and do not fully explain the linkage findings that prompted each study."¹ However, a systematic statistical meta-analytic approach indicates less evidence for DISC1 and RGS4 (see Table 2) and supports other associations. (For a database of all published studies on gene associations with schizophrenia, see www.schizophreniaforum.org/res/sczgene/default.asp).

It is not clear if a gene-environment interaction exists, or if the environmental events act independently. One recent report indicates a specific gene-environment interaction in schizophrenia, in which traumatic brain injury was more likely to be associated with schizophrenia in patients with a family history of that disorder.² In the past the interpersonal environment of the family was thought to play a role in susceptibility, but this does not appear to be the case. Nonetheless, some differences in expressed emotion within families are associated with differences in severity of illness in the patients.

The second half of the 20th century also saw the development of biochemical hypotheses for schizophrenia. The most prominent was the dopamine hypothesis that Carlsson and Lindqvist originally put forth in 1963; it was based on the biochemical effects of treatment with the first antipsychotic drugs, which had been introduced only in 1952. Animal studies lent support to what became a building block of antipsychotic drug discovery - the search for dopamine receptor blockers - and treatment, and Carlsson shared a Nobel Prize for this and other research in 2002.

At this time multiple classes of dopamine receptors are known. Genetic mutations in dopamine receptors have also been discovered and have been applied to comparisons of patients and controls for decades. At first there appeared to be no association between these genetic changes and susceptibility to schizophrenia, but as evidence accumulated from studies of thousands of normal controls and patients with schizophrenia, a pattern emerged. Examination and reanalysis of meta-analyses has implicated three dopamine receptors that have genetic variants consistently associated with schizophrenia: DRD1, DRD2, and DRD4 (dopamine receptors 1, 2, and 4). Other evidence also implicates DRD3. Other specific genes, based on biochemical hypotheses, also have strong evidence for association with schizophrenia.

The overall picture for schizophrenia is now quite promising. The genes associated with schizophrenia vulnerability are also related to treatment in the case of dopamine receptors. This suggests that treatments might be tailored to the susceptibility variants, and that the tailored treatments would be more effective in patients who had the specific variants. It also suggests that new treatment approaches might be developed from translational research based on the more recently developed gene-disease associations. Finally, it invites research on prevention based on precautionary care of women through deterrence of viral infections during pregnancy, and improvements in obstetric delivery.

Elliot S. Gershon is Foundations Fund Professor of Psychiatry and Human Genetics at the University of Chicago.

Table 1. Environmental traumas associated with schizophrenia

Table 2. Genes statistically associated with schizophrenia susceptibility